CN109977322B - Travel mode recommendation method and device, computer equipment and readable storage medium - Google Patents
Travel mode recommendation method and device, computer equipment and readable storage medium Download PDFInfo
- Publication number
- CN109977322B CN109977322B CN201910164479.5A CN201910164479A CN109977322B CN 109977322 B CN109977322 B CN 109977322B CN 201910164479 A CN201910164479 A CN 201910164479A CN 109977322 B CN109977322 B CN 109977322B
- Authority
- CN
- China
- Prior art keywords
- point
- vector
- starting
- starting point
- end point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The application provides a travel mode recommendation method, a travel mode recommendation device, computer equipment and a readable storage medium, wherein the method comprises the following steps: acquiring a starting point and an end point selected by a trip user; determining a starting point representation vector of the starting point and an end point representation vector of the end point according to POI (point of interest) distribution information of the starting point and POI distribution information of the end point; inputting the starting point characterization vector and the end point characterization vector into the trained model to obtain starting point and end point vector characterizations corresponding to the starting point and the end point, wherein the starting point and end point characterization vector is used for indicating travel preference of the corresponding starting point and end point; and obtaining the similarity between the starting and ending point vector representation and the vector representations of the preset multiple travel modes, and determining the target travel mode from the preset multiple travel modes according to the similarity. The method can realize the distribution situation of the interest points based on the starting point and the end point, predict the travel mode preference of the travel user and improve the accuracy of travel mode recommendation.
Description
Technical Field
The present application relates to the field of internet technologies, and in particular, to a travel mode recommendation method and apparatus, a computer device, and a readable storage medium.
Background
With the continuous development of internet technology and the popularization of terminal devices, when a user goes out, a user can input a starting point and a finishing point of the user by using a search Application (APP for short), a map APP and the like in the terminal device, and then the APP can recommend a preferred traffic mode of the user, such as walking, public transportation, driving and the like.
In the prior art, the travel preference of a user is generally predicted and recommended according to the historical travel information of the user. However, this recommendation method is not accurate, and the accuracy of the prediction result is low when the historical trip information is too little.
Disclosure of Invention
The application provides a travel mode recommendation method, a travel mode recommendation device, computer equipment and a readable storage medium, so that travel mode preference of a travel user is predicted and obtained based on the distribution conditions of interest points of a starting point and an end point, and accuracy of travel mode recommendation is improved, and the problem that in the prior art, the prediction result is inaccurate under the condition that the travel mode preference of the user is predicted according to historical travel information of the user and the historical travel information is too little is solved.
An embodiment of a first aspect of the present application provides a travel mode recommendation method, including:
acquiring a starting point and an end point selected by a trip user;
determining a starting point representation vector of the starting point and an end point representation vector of the end point according to the POI distribution information of the interest point of the starting point and the POI distribution information of the end point;
inputting the starting point characterization vector and the end point characterization vector into a trained model to obtain starting point and end point vector characterizations corresponding to the starting point and the end point, wherein the model learns to obtain a corresponding relation between the starting point characterization vector and the end point characterization vector and the starting point and end point vector characterizations, and the starting point and end point characterization vector is used for indicating travel preference of the corresponding starting point and end point;
and obtaining the similarity between the starting and ending point vector representation and vector representations of a plurality of preset travel modes, and determining a target travel mode from the plurality of preset travel modes according to the similarity.
According to the travel mode recommendation method, the starting point characterization vector of the starting point and the end point characterization vector of the end point are determined by obtaining the starting point and the end point selected by a travel user, according to the POI distribution information of the interest point of the starting point and the POI distribution information of the end point, the starting point characterization vector and the end point characterization vector are input into a trained model, and starting and end point vector characterizations corresponding to the starting point and the end point are obtained, wherein the model learns to obtain the corresponding relation between the starting point characterization vector and the end point characterization vector and the starting and end point vector characterizations, the starting and end point characterization vector is used for indicating travel preference of the corresponding starting and end point, and then the similarity between the starting and end point vector characterizations and vector characterizations of a plurality of preset travel modes is obtained, and a target travel mode is determined from the plurality of preset. Therefore, the travel mode preference of the travel user is obtained through prediction based on the distribution conditions of the interest points of the starting point and the end point, and the accuracy of travel mode recommendation is improved, so that the problem that the prediction result is inaccurate under the condition that the historical travel information is too little due to the fact that the travel mode preference of the user is predicted according to the historical travel information of the user in the prior art is solved.
An embodiment of a second aspect of the present application provides a travel mode recommendation device, including:
the system comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring a starting point and an end point selected by a trip user;
the determining module is used for determining a starting point representation vector of the starting point and an end point representation vector of the end point according to the POI distribution information of the starting point and the POI distribution information of the end point;
the processing module is used for inputting the starting point characterization vector and the end point characterization vector into a trained model to obtain starting point and end point vector characterizations corresponding to the starting point and the end point, wherein the model learns to obtain corresponding relations between the starting point characterization vector and the end point characterization vector and between the starting point and the end point vector characterizations, and the starting point and end point characterization vector is used for indicating travel preference of the corresponding starting point and end point;
and the recommending module is used for acquiring the similarity between the starting and ending point vector representation and vector representations of a plurality of preset travel modes, and determining a target travel mode from the plurality of preset travel modes according to the similarity.
The travel mode recommendation device of the embodiment of the application determines a starting point characterization vector of the starting point and an end point characterization vector of the end point by obtaining the starting point and the end point selected by a travel user and inputting the starting point characterization vector and the end point characterization vector into a trained model according to POI distribution information of the interest point of the starting point and POI distribution information of the end point, so as to obtain starting and end point vector characterizations corresponding to the starting point and the end point, wherein the model learns to obtain a corresponding relation between the starting point characterization vector and the end point characterization vector and the starting and end point vector characterizations, and the starting and end point characterization vector is used for indicating travel preference of a corresponding starting and end point, so as to obtain similarity between the starting and end point vector characterizations and vector characterizations of a plurality of preset travel modes, and determine a target travel mode from the plurality of preset travel modes. Therefore, the travel mode preference of the travel user is obtained through prediction based on the distribution conditions of the interest points of the starting point and the end point, and the accuracy of travel mode recommendation is improved, so that the problem that the prediction result is inaccurate under the condition that the historical travel information is too little due to the fact that the travel mode preference of the user is predicted according to the historical travel information of the user in the prior art is solved.
An embodiment of a third aspect of the present application provides a computer device, including: the travel mode recommendation method includes, when the processor executes the program, implementing the travel mode recommendation method provided in the embodiment of the first aspect of the present application.
In order to achieve the above object, a fourth aspect of the present application provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a travel mode recommendation method as set forth in the first aspect of the present application.
Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a schematic flow chart of a travel mode recommendation method according to a first embodiment of the present application;
fig. 2 is a schematic flow chart of a travel mode recommendation method provided in the second embodiment of the present application;
fig. 3 is a schematic flow chart of a travel mode recommendation method provided in the third embodiment of the present application;
fig. 4 is a schematic structural diagram of a travel mode recommendation device according to a first embodiment of the present application;
fig. 5 is a schematic structural diagram of a travel mode recommendation device according to a second embodiment of the present application;
fig. 6 is a schematic structural diagram of a travel mode recommendation device according to a third embodiment of the present application;
FIG. 7 illustrates a block diagram of an exemplary computer device suitable for use to implement embodiments of the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present application and should not be construed as limiting the present application.
A travel mode recommendation method, an apparatus, a computer device, and a readable storage medium according to embodiments of the present application are described below with reference to the accompanying drawings.
In the related art, when travel mode recommendation is performed, a graph is generally established according to historical travel events of a user, vector characteristics of the user, the travel mode and a starting and ending point are learned by using a graph embedding method, and then the travel mode with the highest score is screened out by using an online recommendation method for recommendation.
The recommendation method adopts a graph embedding mode, carries out travel event modeling based on historical travel events, and does not consider the environmental context of the starting point and the ending point. However, during research, the inventor finds that the environmental context has a great influence on the user's travel pattern selection, for example, the user may drive a car from home to a company, and drive a car from one shopping mall to another shopping mall. Therefore, aiming at the problems in the prior art, the application provides the occurrence mode recommendation method, so that the travel mode preference of the user is predicted to occur based on the environment context, and the accuracy of travel mode recommendation is improved.
Since the distribution of the points of Interest (POI) of an area can embody the function of the area, for example, the area where a company or a restaurant is concentrated may be a business area, the area where a park or an attraction is concentrated may be a travel culture area, and so on, the distribution of the POI of the starting Point and the ending Point can be used to represent the environmental context of the starting Point and the ending Point. The travel mode recommendation method is travel mode prediction achieved based on POI distribution information of starting and ending points.
Fig. 1 is a flowchart illustrating a travel mode recommendation method according to an embodiment of the present application.
The embodiment of the present application is exemplified by the travel mode recommendation method being configured in the travel mode recommendation apparatus provided in the present application, and the travel mode recommendation apparatus may be applied to any computer device, so that the computer device may perform a travel mode recommendation function.
The Computer device may be a Personal Computer (PC), a cloud device, a mobile device, and the like, and the mobile device may be a hardware device having various operating systems, touch screens, and/or display screens, such as a mobile phone, a tablet Computer, a Personal digital assistant, a wearable device, and an in-vehicle device.
As shown in fig. 1, the travel mode recommendation method includes the following steps:
In this embodiment, the trip user may input a starting point and an ending point of a trip by searching for a APP, a map APP, and the like, where the manner for the user to input the starting point and the ending point of the trip includes, but is not limited to, touch input (such as sliding, clicking, and the like), keyboard input, voice input, and the like.
In this embodiment, after the travel user inputs the start point and the end point, the travel mode recommendation device may obtain the start point and the end point input by the travel user.
And step 102, determining a starting point characterization vector of the starting point and an end point characterization vector of the end point according to the POI distribution information of the interest point of the starting point and the POI distribution information of the end point.
In this embodiment, after the travel mode recommending apparatus obtains the starting point and the ending point input by the travel user, the POI distribution information corresponding to the starting point region and the POI distribution information corresponding to the ending point region may be respectively obtained according to the regions where the obtained starting point and ending point are located. And determining a starting point representation vector corresponding to the starting point according to the POI distribution information of the starting point, and determining an end point representation vector corresponding to the end point according to the POI distribution information of the end point.
As an example, after acquiring the start point and the end point, the travel mode recommendation apparatus may acquire, from a POI database in which all currently collected POI data are stored, as the POI distribution information of the start point, all POIs whose longitude and latitude difference values do not exceed preset values are acquired, with the start point and the end point respectively serving as centers, where each POI includes information about four aspects of name, type, longitude and latitude. When the POI distribution information of the starting point is obtained, the POI of which the longitude information is in the range of (the starting point longitude value-longitude preset value, the starting point longitude value + longitude preset value) and the latitude information is in the range of (the starting point latitude value-latitude preset value, the starting point latitude value + latitude preset value) is obtained from the POI database as the started POI distribution. Likewise, the POI distribution information of the destination can be determined in a similar manner.
It should be noted that, when acquiring the POI distribution information of the end point, the adopted preset longitude value and preset latitude value may be the same as or different from the preset longitude value and preset latitude value adopted when acquiring the POI distribution information of the start point, which is not limited in this application.
Since the POIs stored in the POI database include types corresponding to the POIs, according to the POI distribution information of the starting point, the starting point characterization vector of the starting point can be determined based on the number of each type of POI included in the POI distribution information corresponding to the starting point, and according to the POI distribution information of the end point, the end point characterization vector of the end point can be determined based on the number of each type of POI included in the POI distribution information corresponding to the end point.
Among them, the type of POI may be, but is not limited to, a primary category including dining (code 01), shopping (code 02), lodging (code 03), going (code 04), cultural and sports entertainment (code 05), financial services (code 06), life services (code 07), automobile services (code 08), education (code 09), medical treatment (code 10), property (code 11), travel (code 12), enterprise and public institution (code 13), administrative institution (code 14), and public service (code 15). When the POI is type-marked, the POI may be marked by a name of the type (for example, a restaurant), or may be marked by a code corresponding to the type (for example, 01 represents a restaurant), which is not limited in this application.
Specifically, the starting point token vector and the ending point token vector may be represented in the form of the following equation (1):
P＝[p1,p2,…,pr] (1)
wherein p iskAnd r is the total number of POI types and is the number of k-th POIs contained in the POI distribution information corresponding to the starting point.
Taking the starting point characterization vector for determining the starting point as an example, as can be seen from the above description of the POI types, when the POIs are classified according to the primary classification, the total number of the POI types is 15, and it is assumed that the POI distribution information of the starting point includes 20 POIs, where the number of POIs of which types are shopping is 9, the number of POIs of which types are traveling is 1, the number of POIs of which types are financial services is 3, the number of POIs of which types are life services is 3, the number of POIs of which types are medical services is 1, the number of POIs of which types are property of property0＝[0 9 0 1 0 3 3 0 0 1 2 0 0 0 1]。
And 103, inputting the starting point characterization vector and the end point characterization vector into the trained model to obtain starting point and end point vector characterizations corresponding to the starting point and the end point, wherein the model learns to obtain the corresponding relation between the starting point characterization vector and the end point characterization vector and the starting point and end point vector characterizations, and the starting point and end point characterization vector is used for indicating the travel preference of the corresponding starting point and end point.
In this embodiment, after the start point characterization vector and the end point characterization vector are determined, the start point characterization vector and the end point characterization vector may be input into a pre-trained model to obtain start point and end point vector characterizations corresponding to the start point and the end point.
Wherein the model is trained in advance. The method comprises the steps that a plurality of pairs of starting and ending points are obtained, training samples containing the starting and ending points can be obtained for improving the accuracy of travel mode recommendation, the model is trained under the constraint condition of the POI distribution information corresponding to the starting and ending points, the trained model is made to learn to obtain corresponding relations between starting point characterization vectors and ending point characterization vectors and starting and ending point vector characterizations, wherein the starting and ending points with similar starting point characterization vectors and ending point characterization vectors correspond to similar vector characterizations, and the starting and ending point characterization vectors are used for indicating travel preference of corresponding starting and ending points.
That is to say, under the constraint of the POI distribution information, the model is trained, so that the starting and ending point vector representation learned by the model can reflect the travel preference of the starting and ending point, and the POI distribution information can remind the function of the region, so that the learned starting and ending point vector representation can also reflect the functional characteristics of the region, the starting and ending point vector representations learned by the starting and ending points with similar functions have larger similarity, and the starting and ending point vector representations learned by the regions with large function difference have smaller similarity.
It should be noted that the process of training the model will be given in the following, and will not be described too much here.
In this embodiment, since the model has learned the corresponding relationship between the start point characterization vector and the end point characterization vector and the start point and end point vector characterization, the determined start point characterization vector and end point characterization vector may be input to the model to obtain the corresponding start point and end point vector characterization.
And 104, acquiring similarity between the starting and ending point characterization vectors and vector characterizations of a plurality of preset travel modes, and determining a target travel mode from the plurality of preset travel modes according to the similarity.
The vector representations of the multiple travel modes may be preset.
In this embodiment, after determining starting and ending point vector representations corresponding to a starting point and an ending point input by a trip user, similarity between the starting and ending point vector representations and vector representations of each preset trip mode may be calculated, where the similarity may be represented by an euclidean distance, a manhattan distance, a cosine similarity, a minkowski distance, and the like, which is not limited in this application. For example, the distance between the start and end point vector representation and the travel mode vector representation can be calculated according to the euclidean distance formula as the similarity between the start and end point vector representation and the travel mode vector representation.
It can be understood that, the higher the similarity between the starting and ending point vector representation and the vector representation of a certain travel mode, the greater the probability that the travel user selects the travel mode from the starting point to the terminal, so that, in this embodiment, after the similarity between the starting and ending point vector representation and the vector representation of each travel mode is obtained, the travel mode corresponding to the maximum similarity may be determined as the target travel mode. Furthermore, the travel mode recommending device can recommend the determined target travel mode to the user through text, voice and other modes.
The travel mode recommendation method of the embodiment determines a starting point characterization vector of the starting point and an end point characterization vector of the end point by obtaining the starting point and the end point selected by a travel user, according to the POI distribution information of the interest point of the starting point and the POI distribution information of the end point, and inputs the starting point characterization vector and the end point characterization vector into a trained model to obtain starting and end point vector characterizations corresponding to the starting point and the end point, wherein the model learns to obtain a corresponding relation between the starting point characterization vector and the end point characterization vector and between the starting and end point vector characterizations, and the starting and end point characterization vector is used for indicating travel preference of a corresponding starting and end point, so as to obtain similarity between the starting and end point vector characterizations and vector characterizations of a plurality of preset travel modes, and determine a target travel mode from the plurality of preset. Therefore, the travel mode preference of the travel user is obtained through prediction based on the distribution conditions of the interest points of the starting point and the end point, and the accuracy of travel mode recommendation is improved, so that the problem that the prediction result is inaccurate under the condition that the historical travel information is too little due to the fact that the travel mode preference of the user is predicted according to the historical travel information of the user in the prior art is solved.
In order to describe more clearly the specific implementation process of determining the starting point representation vector of the starting point and the ending point representation vector of the ending point according to the POI distribution information of the starting point and the POI distribution information of the ending point in the foregoing embodiment, the following description is made in detail with reference to fig. 2, and fig. 2 is a schematic flow chart of the travel mode recommendation method provided in the second embodiment of the present application.
As shown in fig. 2, based on the embodiment shown in fig. 1, step 102 may include the following steps:
step 201, a first POI set of all POIs included in a circular area formed by taking a starting point as a circle center and a first preset distance as a radius is obtained, and a second POI set of all POIs included in the circular area formed by taking an end point as the circle center and a second preset distance as the radius is obtained.
In this embodiment, after the starting point and the end point input by the travel user are obtained, a circle may be drawn with the starting point as a center of the circle and the first preset distance as a radius to form a circular area with the starting point as a center, all POIs included in the circular area are obtained to form a first POI set, the end point is obtained as a center of the circle and the first preset distance is used as a radius to draw a circle to form a circular area with the end point as a center, and all POIs included in the circular area are obtained to form a second POI set, where the first preset distance and the second preset distance may be the same or different, which is not limited in this application.
In this embodiment, after the first POI set corresponding to the starting point is obtained, the type corresponding to each POI may be determined for each POI included in the first POI set.
As an example, the name of the POI may be input into a classification model trained in advance to obtain a type matching the POI.
As an example, correspondence between different POIs and types may be stored in advance, and then the POIs in the first POI set may be compared with the POIs in the pre-stored correspondence to determine the corresponding types.
Then, after the type of each POI in the first POI set is determined, the number of each type of POI can be counted, and then a starting point characterization vector is generated according to the counting result. For example, the starting point characterization vector may be represented according to the aforementioned formula (1). If the first POI set does not include a type of POI, the element corresponding to the number of POIs of the type in the starting point characterization vector is 0.
The number of POI types refers to the total number of categories of the current existing POI classification, but not the number of POI types included in the first POI set or the second POI set, for example, the POI is currently subjected to primary classification, all POIs are divided into 15 types, the number of POI types is 15, and the dimensions of the start point characterization vector and the end point characterization vector are also 15.
It should be noted that, according to the number of each type of POI included in the second POI set, the process of generating the end point token vector is similar to the process of generating the start point token vector, and is not described herein again.
According to the travel mode recommendation method, the first POI set of all POIs contained in the circular area formed by taking the starting point as the circle center and the first preset distance as the radius is obtained, the second POI set of all POIs contained in the circular area formed by taking the end point as the circle center and the second preset distance as the radius is obtained, the starting point characterization vector is generated according to the number of each type of POI contained in the first POI set, and the end point characterization vector is generated according to the number of each type of POI contained in the second POI set, so that a foundation is laid for realizing the prediction of the occurrence mode based on the distribution condition of the POIs.
Although the trip mode predicted according to the environmental context information of the starting point and the ending point is more accurate, due to the individual particularity of the trip user and the noise of the environmental context information, the trip mode is predicted only according to the environmental context information, and the individual preference of the trip user is not considered, so that the trip mode is one-sided and unreasonable. For example, when the distance between the starting point and the ending point is 5-10 km, a trip user with a common income consumption level may choose to ride, transit or drive from the starting point to the ending point, and the trip mode of the trip user cannot be accurately predicted only according to the environmental context information. Therefore, in order to further improve the accuracy of the travel mode recommendation, in a possible implementation manner of the embodiment of the application, the historical travel mode record of the travel user may be considered, that is, the travel preference of the travel user is determined comprehensively according to the historical travel mode record of the travel user and the environmental context information of the starting and ending point. Therefore, during model training, historical travel mode records of a travel user can be used as a main frame of the model, POI distribution information of the starting point and the ending point is used as supplementary information to refine the model, vector representations learned by the model can contain personalized semantics, and the problem of prediction distortion existing in travel preference prediction only according to the POI distribution information of the starting point and the ending point can be effectively solved. The training samples comprise historical travel mode records corresponding to each pair of starting and ending points. The above process is described in detail below with reference to fig. 3.
Fig. 3 is a flowchart illustrating a travel mode recommendation method provided in the third embodiment of the present application.
As shown in fig. 3, training the model under the constraint condition of multiple pairs of POI distribution information corresponding to the start and end points may specifically include the following steps:
In this embodiment, for each pair of start and end points included in the training sample, POI distribution information of the start point and POI distribution information of the end point may be obtained, and then, according to the POI distribution information of the start point and the POI distribution information of the end point, a vector representation corresponding to each pair of start and end points is determined. For a manner of obtaining the POI distribution information of the starting point and the terminal, reference may be made to the description of obtaining the POI distribution information in the foregoing embodiment, which is not described herein again.
Specifically, the starting point characterization vector P of the starting point can be determined according to the POI distribution information of the starting pointoAnd determining an end point representation vector P of the end point according to the POI distribution information of the terminaldWherein P isoAnd PdExpressed in the form of equation (1) with dimensions equal to the number of POI types. And then representing the vector representation of the starting point and the ending point as a starting point representation vector PoAnd endpoint characterization vector PdIs shown in equation (2).
Wherein the content of the first and second substances,
represents a cascade, PodRepresenting the vector representation, P, corresponding to the starting and ending pointsodHas a dimension of PoAnd PdIs the sum of the dimensions of, i.e., if PoAnd PdIs r, then PodHas a dimension of 2 r.
Because the distance between the starting point and the ending point has a certain influence on the travel mode selected by the user, for example, the user traveling within 2 kilometers is likely to select walking, and the possibility of selecting walking over 10 kilometers is very low, the distance between the starting point and the ending point can also be used as the environmental context information of the starting point and the ending point to represent the characteristics of the starting point and the ending point. Therefore, in a possible implementation manner of the embodiment of the application, when the vector characterization corresponding to the start and end points is determined, the distance between the start and end points can be taken into consideration to more accurately characterize the start and end points.
Therefore, according to the POI distribution information corresponding to the multiple pairs of starting and ending points, determining the vector representation corresponding to each pair of starting and ending points in the training sample comprises the following steps: acquiring a spherical distance between each pair of start and end points in a training sample; determining a starting point characterization vector and an end point characterization vector corresponding to each pair of starting and ending points according to the POI distribution information of each pair of starting and ending points; and generating vector representations corresponding to the starting point and the ending point according to the spherical distance, the starting point representation vector and the ending point representation vector corresponding to each pair of starting point and ending point.
For each pair of start and end points in the training sample, the spherical distance between the start and end points may be calculated according to the longitude and latitude of the start and end points, for example, the spherical distance may be calculated according to formula (3):
dod＝R*arccos(sin(x1)*sin(x2)+cos(x1)*cos(x2)*cos(y1-y2)) (3)
wherein od denotes start and end points, dodIndicating the spherical distance, x, between the starting and ending points1Indicates the latitude of the origin, y1Longitude, x, representing the origin2Latitude, y, representing the endpoint2Represents the longitude of the endpoint, and R is the earth's radius.
For each pair of start and end points in the training sample, the determined spherical distance d is usedodStarting point characterization vector PoAnd endpoint characterization vector PdThen the vector representation P corresponding to the starting point and the ending point can be determinedodThe calculation formula is shown as formula (4):
wherein, in the formula (4), dodIs considered as a one-dimensional vector, PodHas a dimension of (1+2r), wherein r represents PoAnd PdOf (c) is calculated.
When the travel frequency is less in the history travel mode record, the travel preference of the travel user cannot be embodied, for example, the travel user only uses the map APP once when traveling, and clicks the bus, and at the moment, the travel preference of the travel user cannot be determined to be the bus. Therefore, in a possible implementation manner of the embodiment of the application, in order to improve the accuracy of the model prediction result, the start and end points included in the training sample may be filtered start and end points, for example, only the start and end points with the trip frequency reaching a preset threshold (for example, 20 times) are used as the training sample, that is, in the historical trip mode record corresponding to each pair of start and end points in the training sample, the trip frequency reaches the preset threshold.
And 302, determining the travel preference vector representation corresponding to the starting point and the ending point according to the historical travel mode record.
In this embodiment, according to the historical travel mode record included in the training sample and corresponding to each pair of start and end points, the travel preference vector representation of each pair of start and end points can be determined.
As an example, for a certain starting and ending point, the number of times or frequency of each travel mode in the historical travel mode record corresponding to the starting and ending point may be counted, and then according to the statistical result, the travel preference vector of the starting and ending point is represented in the form of formula (5).
Wherein odiIndicating the ith starting and ending point in the training sample,
representing the travel preference vector representation of the ith starting and ending point, mkAnd (k is 1, 2, …, n) represents the frequency of the k-th travel mode in the historical travel house record corresponding to the i-th starting and ending point, and n is the total number of the travel mode types.
Due to the fact that the POIs of different types have different influence degrees on the preference of the travel mode, different weights can be distributed to different POI types in the embodiment, and accuracy of the prediction result is improved.
As an example, an initial weight may be assigned to each POI type, a vector representation of the same starting point and ending point may be matched with a travel preference vector representation as a target, and based on a logistic regression model, the weight of the POI type may be continuously adjusted for learning, so as to finally obtain the weight corresponding to different POI types. The method comprises the steps of representing weights of different POI types into a vector form, taking a random vector as an initialized weight, calculating an inner product of the weights and vector representations of a starting point and a finishing point, executing a multi-classification task by using a calculation result, continuously fitting distribution of travel preference vector representations of the starting point and the finishing point, and adjusting values of the weights in each iteration so as to enable the degree of fitting of the adjusted weights with the inner product of the vector representations of the starting point and the finishing point and the travel preference vector representations to be maximum.
And step 304, determining the travel preference correlation between different starting points and different ending points according to the adjusted weight.
In this embodiment, after learning to obtain the weights of different POI types, the travel preference correlations between different start points and different end points may be determined according to the adjusted weights.
As an example, the travel preference correlation between different starting and ending points can be calculated according to the following formula (6).
rel(odi,odj)＝exp(-||W⊙(odi-odj)||) (6)
Wherein, odiAnd odjRepresenting two different start and end points and W representing the weights of the different POI types.
And 305, generating an objective function according to the correlation among the vector representations corresponding to different starting and ending points and the travel preference correlation corresponding to the starting and ending points.
As an example, historical behavior data of different users may be obtained, and a graph embedding model may be constructed according to the historical behavior data, where a user, a transportation mode, and a start and end point represent nodes in a graph corresponding to the graph embedding model, and an edge between two nodes represents a correlation between the nodes. The first objective function corresponding to the graph embedding model is described in equation (7):
wherein the content of the first and second substances,
vector characterization, x, representing users whose models need to be learnedmVector representation of the traffic pattern corresponding to the positive sample, x, to be learned by the modelm'The vector representation of the traffic mode corresponding to the negative sample required to be learned by the model,
vector representation of the starting and ending points of the model to be learned, epsilonumRepresenting the edge, ε, between the user and the mode of transportation in the graphodmThe side between the starting and ending point and the transportation means in the figure is shown, and U represents the transportation means set.
Taking the environmental context information of the starting point and the ending point as a supplementary condition, generating a second objective function as shown in formula (8):
wherein the content of the first and second substances,
and representing the correlation between the vector representations corresponding to the starting and ending points i and j.
For any edge, the gradient of the environmental context part is:
wherein the content of the first and second substances,
and
vector characterization for respectively representing start and end points needing learning, alpha represents learning rate, beta1Representing personalized item weights. The gradient is mainly used for vector characterization of each node in the graph corresponding to the iterative optimization graph embedding model in training.
Then, the final generated objective function is shown in equation (10):
O＝O0+δ*O1 (10)
thus, by introducing the second objective function O1The environmental context information is used as a regularization item of the model to carry out model training, so that the vector representations learned by the starting points and the ending points close to the environmental context information have higher similarity, andthe vector representations learned by the starting point and the ending point with larger difference of the environmental context information have lower similarity.
The model is trained to minimize the objective function, step 306.
In this embodiment, after the target function is generated, the target function may be minimized, and the model is trained by using the training samples to obtain a trained model for predicting the start-stop point vector representation.
According to the travel mode recommendation method, the influence of the environment context information of the starting point and the ending point on the starting point and the ending point representation is considered during model training, so that the accuracy of the starting point and the ending point representation can be improved, and the accuracy of travel mode recommendation is further improved.
The inventor tests the travel mode recommendation method (marked as algorithm 1) proposed in the present application, the existing graph embedding model recommendation method (marked as algorithm 2), the ranking model recommendation method (marked as algorithm 3) and the logistic regression recommendation method (marked as algorithm 4) on the Beijing and Shanghai datasets, and the test results are shown in Table 1.
TABLE 1
Where ndcg @5 represents the normalized loss accumulation gain.
As can be seen from table 1, the result of the travel mode recommendation method provided by the application on each evaluation index of ndcg @5, accuracy, recall rate and F value is mostly superior to that of the existing recommendation mode.
In order to implement the above embodiment, the present application further provides a travel mode recommendation device.
Fig. 4 is a schematic structural diagram of a travel mode recommendation device according to a first embodiment of the present application.
As shown in fig. 4, the travel mode recommendation device 40 includes: an acquisition module 410, a determination module 420, a processing module 430, and a recommendation module 440.
The obtaining module 410 is configured to obtain a starting point and an ending point selected by the trip user.
The determining module 420 is configured to determine a starting point characterization vector of the starting point and an end point characterization vector of the end point according to the POI distribution information of the starting point and the POI distribution information of the end point.
And the processing module 430 is configured to input the starting point characterization vector and the end point characterization vector into the trained model, so as to obtain starting and end point vector characterizations corresponding to the starting point and the end point, where the model learns the corresponding relationship between the starting point characterization vector and the end point characterization vector and the starting and end point vector characterizations, and the starting and end point characterization vector is used for indicating travel preference of the corresponding starting and end point.
And the recommending module 440 is configured to obtain similarity between the starting and ending point vector characterization and vector characterization of a plurality of preset travel modes, and determine a target travel mode from the plurality of preset travel modes according to the similarity.
Further, in a possible implementation manner of the embodiment of the present application, as shown in fig. 5, on the basis of the embodiment shown in fig. 4, the determining module 420 includes:
the obtaining unit 421 is configured to obtain a first POI set of all POIs included in a circular area formed by taking the starting point as a circle center and the first preset distance as a radius, and obtain a second POI set of all POIs included in the circular area formed by taking the end point as the circle center and the second preset distance as a radius.
A determining unit 422, configured to generate a starting point characterization vector according to the number of each type of POI included in the first POI set; generating an end point characterization vector according to the number of each type of POI contained in the second POI set; and the dimensions of the starting point characterization vector and the end point characterization vector are the same as the number of POI types.
In a possible implementation manner of the embodiment of the present application, as shown in fig. 6, on the basis of the embodiment shown in fig. 4, the travel mode recommending apparatus 40 further includes:
the training module 400 is configured to obtain a training sample, where the training sample includes multiple pairs of start and end points, and train the model under the constraint condition of POI distribution information corresponding to the multiple pairs of start and end points.
In a possible implementation manner of the embodiment of the present application, the training module 400 is specifically configured to: determining vector representations corresponding to each pair of start and end points in a training sample according to POI distribution information corresponding to the plurality of pairs of start and end points; determining travel preference vector representations corresponding to starting and ending points according to historical travel mode records; carrying out weight adjustment on different POI types so as to enable the vector representation of the same starting point and the same ending point to be matched with the travel preference vector representation; determining the travel preference correlation between different starting points and different ending points according to the adjusted weight; generating an objective function according to the correlation among the vector representations corresponding to different starting and ending points and the trip preference correlation corresponding to the starting and ending points; the model is trained to minimize the objective function.
As an example, when determining the travel preference correlation between different starting and ending points, the travel preference correlation may be determined according to the following formula:
rel(odi,odj)＝exp(-||W⊙(odi-odj)||)；
wherein odiAnd odjRepresenting two different start and end points and W representing the weights of the different POI types.
In a possible implementation manner of the embodiment of the present application, when determining the vector characterization corresponding to each pair of start and end points in the training sample according to the POI distribution information corresponding to the plurality of pairs of start and end points, the training model 400 is configured to: acquiring a spherical distance between each pair of start and end points in a training sample; determining a starting point characterization vector and an end point characterization vector corresponding to each pair of starting and ending points according to the POI distribution information of each pair of starting and ending points; and generating vector representations corresponding to the starting point and the ending point according to the spherical distance, the starting point representation vector and the ending point representation vector corresponding to each pair of starting point and ending point.
It should be noted that the foregoing explanation of the embodiment of the travel method recommendation method is also applicable to the travel method recommendation device of the embodiment, and the implementation principle thereof is similar and will not be described herein again.
The travel mode recommendation device of the embodiment of the application determines a starting point characterization vector of the starting point and an end point characterization vector of the end point by obtaining the starting point and the end point selected by a travel user and inputting the starting point characterization vector and the end point characterization vector into a trained model according to POI distribution information of the interest point of the starting point and POI distribution information of the end point, so as to obtain starting and end point vector characterizations corresponding to the starting point and the end point, wherein the model learns to obtain a corresponding relation between the starting point characterization vector and the end point characterization vector and the starting and end point vector characterizations, and the starting and end point characterization vector is used for indicating travel preference of a corresponding starting and end point, so as to obtain similarity between the starting and end point vector characterizations and vector characterizations of a plurality of preset travel modes, and determine a target travel mode from the plurality of preset travel modes. Therefore, the travel mode preference of the travel user is obtained through prediction based on the distribution conditions of the interest points of the starting point and the end point, and the accuracy of travel mode recommendation is improved, so that the problem that the prediction result is inaccurate under the condition that the historical travel information is too little due to the fact that the travel mode preference of the user is predicted according to the historical travel information of the user in the prior art is solved.
In order to implement the foregoing embodiments, the present application also provides a computer device, including: the travel mode recommendation method includes a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the program, the travel mode recommendation method is implemented as proposed in the foregoing embodiments of the present application.
In order to implement the foregoing embodiments, the present application also proposes a non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a travel mode recommendation method as proposed in the foregoing embodiments of the present application.
FIG. 7 illustrates a block diagram of an exemplary computer device suitable for use to implement embodiments of the present application. The computer device 12 shown in fig. 7 is only an example, and should not bring any limitation to the function and the scope of use of the embodiments of the present application.
As shown in FIG. 7, computer device 12 is in the form of a general purpose computing device. The components of computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that couples various system components including the system memory 28 and the processing unit 16.
Computer device 12 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by computer device 12 and includes both volatile and nonvolatile media, removable and non-removable media.
A program/utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. Program modules 42 generally perform the functions and/or methodologies of the embodiments described herein.
Computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), with one or more devices that enable a user to interact with computer device 12, and/or with any devices (e.g., network card, modem, etc.) that enable computer device 12 to communicate with one or more other computing devices. Such communication may be through an input/output (I/O) interface 22. Moreover, computer device 12 may also communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public Network such as the Internet) via Network adapter 20. As shown in FIG. 7, the network adapter 20 communicates with the other modules of the computer device 12 via the bus 18. It should be appreciated that although not shown in FIG. 7, other hardware and/or software modules may be used in conjunction with computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
The processing unit 16 executes various functional applications and data processing, such as implementing the travel mode recommendation method mentioned in the foregoing embodiments, by running a program stored in the system memory 28.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer case (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory. Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a programmable gate array, a field programmable gate array, or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (10)
1. A travel mode recommendation method is characterized by comprising the following steps:
acquiring a starting point and an end point selected by a trip user;
determining a starting point representation vector of the starting point and an end point representation vector of the end point according to the POI distribution information of the interest point of the starting point and the POI distribution information of the end point;
inputting the starting point characterization vector and the end point characterization vector into a trained model to obtain starting point and end point vector characterizations corresponding to the starting point and the end point, wherein the model learns to obtain a corresponding relation between the starting point characterization vector and the end point characterization vector and the starting point and end point vector characterizations, and the starting point and end point characterization vector is used for indicating travel preference of the corresponding starting point and end point;
and obtaining the similarity between the starting and ending point vector representation and vector representations of a plurality of preset travel modes, and determining a target travel mode from the plurality of preset travel modes according to the similarity.
2. A travel mode recommendation method according to claim 1, wherein said determining a starting point representation vector of the starting point and an ending point representation vector of the ending point according to the POI distribution information of the starting point and the POI distribution information of the ending point comprises:
acquiring a first POI set of all POIs contained in a circular area formed by taking the starting point as the circle center and taking the first preset distance as the radius, and acquiring a second POI set of all POIs contained in the circular area formed by taking the end point as the circle center and taking the second preset distance as the radius;
generating the starting point characterization vector according to the number of each type of POI contained in the first POI set;
generating the end point characterization vector according to the number of each type of POI contained in the second POI set; wherein the dimensions of the start point token vector and the end point token vector are the same as the number of POI types.
3. A method of travel recommendation according to claim 1, before said inputting said start point and said end point into a trained model and obtaining a start point and end point vector representation corresponding to said start point and said end point, further comprising:
obtaining a training sample, wherein the training sample comprises a plurality of pairs of starting and ending points;
and training the model under the constraint condition of POI distribution information corresponding to the multiple pairs of starting and ending points.
4. A travel mode recommendation method according to claim 3, wherein the training samples include historical travel mode records corresponding to each pair of start and end points, and the training of the model under the constraint condition of the POI distribution information corresponding to the plurality of pairs of start and end points includes:
determining vector representations corresponding to each pair of start and end points in the training sample according to POI distribution information corresponding to the plurality of pairs of start and end points;
determining travel preference vector representations corresponding to starting and ending points according to the historical travel mode records;
carrying out weight adjustment on different POI types so as to enable the vector representation of the same starting point and the same ending point to be matched with the travel preference vector representation;
determining the travel preference correlation between different starting points and different ending points according to the adjusted weight;
generating an objective function according to the correlation among the vector representations corresponding to different starting and ending points and the trip preference correlation corresponding to the starting and ending points;
the model is trained to minimize the objective function.
5. A travel mode recommendation method according to claim 4, characterized in that the travel preference correlation between different starting and ending points is determined according to the following formula:
rel(odi,odj)＝exp(-||W⊙(odi-odj)||)，
wherein odiAnd odjRepresenting two different start and end points and W representing the weights of the different POI types.
6. A travel mode recommendation method according to claim 4, wherein the determining, according to the POI distribution information corresponding to the multiple pairs of start and end points, the vector characterization corresponding to each pair of start and end points in the training sample comprises:
acquiring a spherical distance between each pair of start and end points in the training sample;
determining a starting point characterization vector and an end point characterization vector corresponding to each pair of starting and ending points according to the POI distribution information of each pair of starting and ending points;
and generating vector representations corresponding to the starting point and the ending point according to the spherical distance, the starting point representation vector and the ending point representation vector corresponding to each pair of starting points and ending points.
7. A travel mode recommendation device, characterized in that the device comprises:
the system comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring a starting point and an end point selected by a trip user;
the determining module is used for determining a starting point representation vector of the starting point and an end point representation vector of the end point according to the POI distribution information of the starting point and the POI distribution information of the end point;
the processing module is used for inputting the starting point characterization vector and the end point characterization vector into a trained model to obtain starting point and end point vector characterizations corresponding to the starting point and the end point, wherein the model learns to obtain corresponding relations between the starting point characterization vector and the end point characterization vector and between the starting point and the end point vector characterizations, and the starting point and end point characterization vector is used for indicating travel preference of the corresponding starting point and end point;
and the recommending module is used for acquiring the similarity between the starting and ending point vector representation and vector representations of a plurality of preset travel modes, and determining a target travel mode from the plurality of preset travel modes according to the similarity.
8. A travel mode recommendation device according to claim 7, wherein said determining means comprises:
the acquisition unit is used for acquiring a first POI set of all POIs contained in a circular area formed by taking the starting point as the circle center and a first preset distance as the radius, and acquiring a second POI set of all POIs contained in the circular area formed by taking the end point as the circle center and a second preset distance as the radius;
a determining unit, configured to generate the starting point characterization vector according to the number of each type of POI included in the first POI set; generating the end point characterization vector according to the number of each type of POI contained in the second POI set; wherein the dimensions of the start point token vector and the end point token vector are the same as the number of POI types.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the travel mode recommendation method according to any one of claims 1-6 when executing the program.
10. A non-transitory computer-readable storage medium having stored thereon a computer program, wherein the program, when executed by a processor, implements a travel mode recommendation method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910164479.5A CN109977322B (en) | 2019-03-05 | 2019-03-05 | Travel mode recommendation method and device, computer equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910164479.5A CN109977322B (en) | 2019-03-05 | 2019-03-05 | Travel mode recommendation method and device, computer equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109977322A CN109977322A (en) | 2019-07-05 |
| CN109977322B true CN109977322B (en) | 2021-03-23 |
Family
ID=67077961
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910164479.5A Active CN109977322B (en) | 2019-03-05 | 2019-03-05 | Travel mode recommendation method and device, computer equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109977322B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110443490A (en) * | 2019-08-01 | 2019-11-12 | 福建亿榕信息技术有限公司 | A kind of safe traveling method and system based on interaction supervision |
| CN111121806B (en) * | 2019-12-12 | 2021-11-09 | 上海钧正网络科技有限公司 | Travel mode planning method and device, computer equipment and storage medium |
| CN111159583B (en) * | 2019-12-31 | 2023-08-04 | 中国联合网络通信集团有限公司 | User behavior analysis method, device, equipment and storage medium |
| CN111340673A (en) * | 2020-01-09 | 2020-06-26 | 北京航空航天大学 | Travel time consumption calculation method based on air-rail coupling network |
| CN111859184A (en) * | 2020-06-28 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | Boarding point recommendation method and system |
| CN112884235B (en) * | 2021-03-04 | 2022-02-22 | 腾讯科技（深圳）有限公司 | Travel recommendation method, and training method and device of travel recommendation model |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8504285B2 (en) * | 2009-05-15 | 2013-08-06 | Nokia Corporation | Method and apparatus of route determination |
| CN104969261A (en) * | 2013-02-04 | 2015-10-07 | 哈曼国际工业有限公司 | Method and system for detecting moving objects |
| CN105354770A (en) * | 2015-11-16 | 2016-02-24 | 南京途牛科技有限公司 | Real-time price comparison method for route type tourism product |
| CN105488597A (en) * | 2015-12-28 | 2016-04-13 | 中国民航信息网络股份有限公司 | Passenger destination prediction method and system |
| CN107490385A (en) * | 2017-08-21 | 2017-12-19 | 百度在线网络技术（北京）有限公司 | Traffic path planing method and its device |
| CN108242203A (en) * | 2016-12-23 | 2018-07-03 | 中兴通讯股份有限公司 | A kind of region partitioning method and device |
| CN108921587A (en) * | 2018-05-24 | 2018-11-30 | 腾讯科技（深圳）有限公司 | A kind of data processing method, device and server |
| CN109359287A (en) * | 2018-07-12 | 2019-02-19 | 福州大学 | The online recommender system of interactive cultural tour scenic area and scenic spot and method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103512581B (en) * | 2012-06-28 | 2016-12-21 | 北京搜狗科技发展有限公司 | A kind of paths planning method and device |
| CN106021305A (en) * | 2016-05-05 | 2016-10-12 | 北京邮电大学 | Mode and preference sensing POI recommendation method and system |
| CN107633317B (en) * | 2017-06-15 | 2021-09-21 | 北京百度网讯科技有限公司 | Method and device for establishing journey planning model and planning journey |
| CN107656987B (en) * | 2017-09-13 | 2020-07-14 | 大连理工大学 | Subway station function mining method based on L DA model |
| CN108062857B (en) * | 2017-12-22 | 2019-08-23 | 西南交通大学 | Prediction technique for cab-getter's trip purpose |
| CN108829744B (en) * | 2018-05-24 | 2021-07-06 | 湖北文理学院 | Travel mode recommendation method based on situation elements and user preferences |
-
2019
- 2019-03-05 CN CN201910164479.5A patent/CN109977322B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8504285B2 (en) * | 2009-05-15 | 2013-08-06 | Nokia Corporation | Method and apparatus of route determination |
| CN104969261A (en) * | 2013-02-04 | 2015-10-07 | 哈曼国际工业有限公司 | Method and system for detecting moving objects |
| CN105354770A (en) * | 2015-11-16 | 2016-02-24 | 南京途牛科技有限公司 | Real-time price comparison method for route type tourism product |
| CN105488597A (en) * | 2015-12-28 | 2016-04-13 | 中国民航信息网络股份有限公司 | Passenger destination prediction method and system |
| CN108242203A (en) * | 2016-12-23 | 2018-07-03 | 中兴通讯股份有限公司 | A kind of region partitioning method and device |
| CN107490385A (en) * | 2017-08-21 | 2017-12-19 | 百度在线网络技术（北京）有限公司 | Traffic path planing method and its device |
| CN108921587A (en) * | 2018-05-24 | 2018-11-30 | 腾讯科技（深圳）有限公司 | A kind of data processing method, device and server |
| CN109359287A (en) * | 2018-07-12 | 2019-02-19 | 福州大学 | The online recommender system of interactive cultural tour scenic area and scenic spot and method |
Non-Patent Citations (1)
| Title |
|---|
| 基于Android平台下的智能聚餐系统的设计与实现;潘兆轩;《中国优秀硕士学位论文全文数据库 信息科技辑（月刊）》;20180615(第6期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109977322A (en) | 2019-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109977322B (en) | Travel mode recommendation method and device, computer equipment and readable storage medium | |
| Fogli et al. | Exploiting semantics for context-aware itinerary recommendation | |
| GB2550523A (en) | Methods and systems for order processing | |
| US20150161529A1 (en) | Identifying Related Events for Event Ticket Network Systems | |
| US10067985B2 (en) | Computing system with crowd-source mechanism and method of operation thereof | |
| CN111767366B (en) | Question and answer resource mining method and device, computer equipment and storage medium | |
| CN112115257A (en) | Method and apparatus for generating information evaluation model | |
| US20220179882A1 (en) | Method, apparatus, and system for combining location data sources | |
| US11741511B2 (en) | Systems and methods of business categorization and service recommendation | |
| CN111061979B (en) | User tag pushing method and device, electronic equipment and medium | |
| US11537844B2 (en) | Systems and methods of business categorization and service recommendation | |
| US8122002B2 (en) | Information processing device, information processing method, and program | |
| CN109783631A (en) | Method of calibration, device, computer equipment and the storage medium of community's question and answer data | |
| CN111222847B (en) | Open source community developer recommendation method based on deep learning and unsupervised clustering | |
| Zhang et al. | Multi-features taxi destination prediction with frequency domain processing | |
| Zhao et al. | Annotating points of interest with geo-tagged tweets | |
| CN113407854A (en) | Application recommendation method, device and equipment and computer readable storage medium | |
| CN116917887A (en) | Query processing using attention-based ranking system | |
| CN116089567A (en) | Recommendation method, device, equipment and storage medium for search keywords | |
| CN111310065A (en) | Social contact recommendation method and device, server and storage medium | |
| CN107944026A (en) | A kind of method, apparatus, server and the storage medium of atlas personalized recommendation | |
| CN112765339B (en) | Personalized book recommendation method and system based on reinforcement learning | |
| CN111191781A (en) | Method of training neural network, object recognition method and apparatus, and medium | |
| US20210295170A1 (en) | Removal of engagement bias in online service | |
| CN116738066A (en) | Rural travel service recommendation method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |